The transparency of the vertebrate lens depends on assembly of connexins into gap junctions that are required for lens ionic homeostasis and metabolite circulation. Mutations in fiber cell connexins (Cx46, Cx50) in humans and mice result in loss of lens transparency and cataracts, which are a major cause of blindness worldwide. In differentiating fiber (DF) cells, connexins assemble into large plaques on fiber cell broad sides, whereas in organelle-free mature fiber (MF) cells, the plaques fragment and disperse around the membranes. A cross-linked network of spectrin and actin filaments (F-actin), attaches to NrCAM and N-cadherin, creating micron-scale membrane subdomains in lens fiber cells. The F-actin linkers are capped by tropomodulin1 (Tmod1), and coated with tropomyosin (?TM), stabilizing spectrin-actin network integrity. We have shown that absence of Tmod1 in the mouse lens leads to F-actin disassembly and spectrin-actin network disruption, along with abnormal cell shapes. Tmod1-/- lenses are transparent, but Tmod2 levels are elevated, suggesting partial compensation. Recently, in collaboration with Rick Mathias, we found that gap junction coupling conductance in Tmod1-/- lenses is reduced to about half normal, consistent with ~2-fold elevations in hydrostatic pressure and intracellular Na+. Total levels of Cx46 and Cx50 are unchanged, but confocal microscopy reveals that the large Cx46 and Cx50 plaques on the broad sides of DF cels are dispersed to smaller puncta in Tmod1-/- lenses. We hypothesize that assembly of a long-range spectrin-actin network linked to its attachment proteins (NrCAM, N-cadherin) forms a fiber cell membrane subdomain that excludes connexins, thereby 'corralling' the connexins into the large gap junction plaques required for optimal fiber cell coupling in DF. We will test this hypothesis using mouse knock-out models to destabilize F-actin and weaken spectrin-actin network linkages (Tmod1-/-, Tmod2-/-, ?TM-/-), and a knock-in model to strengthen network linkages (calpain/caspase-resistant II-spectrin mutant). The Specific Aims are: 1) to investigate the roles of Tmods in Cx46 and Cx50 assembly and function in gap junction plaques of DF cells by analysis of Tmod1-/- and Tmod1-/-;Tmod2-/- lenses. 2) To investigate TM regulation of spectrin-actin network stability, connexin assembly, and gap junction coupling in DF by analysis of ?TM?exon9d-/- lenses. 3) To investigate gap junctions in MF cells with a calpain/caspase- resistant II-spectrin mutant that is expected to prevent normal spectrin-actin network disassembly in MF (with J. Morrow). Fiber cell membrane subdomains containing connexins or the spectrin-actin network will be defined by confocal and TIRF microscopy, protein associations by biochemical approaches, gap junction plaque morphology by freeze-fracture TEM (with W.-K. Lo) and lens ionic homeostasis and coupling by electrophysiology of whole lenses (with R. Mathias). These studies will establish basic mechanisms for gap junction control of lens transparency, and provide new insights for therapies in cataract prevention.